UNIGE document Scientific Article
previous document  unige:22899  next document
add to browser collection
Title

Robust Inference for Generalized Linear Models

Authors
Published in Journal of the American Statistical Association. 2001, vol. 96, no. 455, p. 1022-1030
Abstract By starting from a natural class of robust estimators for generalized linear models based on the notion of quasi-likelihood, we define robust deviances that can be used for stepwise model selection as in the classical framework. We derive the asymptotic distribution of tests based on robust deviances, and we investigate the stability of their asymptotic level under contamination. The binomial and Poisson models are treated in detail. Two applications to real data and a sensitivity analysis show that the inference obtained by means of the new techniques is more reliable than that obtained by classical estimation and testing procedures.
Keywords Binomial regressionInfluence functionM-estimatorsModel selectionPoisson regressionQuasi-likehoodRobust devianceRobustness of efficiencyRobustness of validity
Identifiers
Full text
Structures
Citation
(ISO format)
CANTONI, Eva, RONCHETTI, Elvezio. Robust Inference for Generalized Linear Models. In: Journal of the American Statistical Association, 2001, vol. 96, n° 455, p. 1022-1030. https://archive-ouverte.unige.ch/unige:22899

417 hits

663 downloads

Update

Deposited on : 2012-09-12

Export document
Format :
Citation style :